Optimal Spectral Analysis for detection of sinusitis using Near-Infrared Spectroscopy

Optical recording technique overtakes neural recording technique for simplicity, non-ionization, portability, non-invasiveness and cost effectiveness. Many optical techniques like FMRI, DOT, SPR, ERP and PPG are in use now-a-days. Among them, Near Infrared Spectroscopy become popular because of its high temporal resolution and reasonable spatial resolution. Keeping this in mind, NIRS can be used to detect the brain activities. As an initiative, a prototype head band for NIRS has been developed and it can be used to detect sinusitis in an easy manner. Signals have been captured from various persons and their statistical parameters have been analyzed. Signals have been captured before and after sleep for normal and sinusitis persons and the difference between them in various statistical parameters were clearly identified. Spectral analysis was done using parametric and non-parametric methods and the best method was identified for detecting sinusitis. Best method is used to analyze normal and sinusitis persons and the increase in peak value and dynamic range for sinusitis person is identified from the waveform